14 Aug
2015
14 Aug
'15
6:03 p.m.
On Sat, Aug 08, 2015 at 01:06:20AM +0530, Subhransu S. Prusty wrote:
- struct skl_pipeline *ppl;
- pipe = devm_kzalloc(dev, sizeof(*pipe), GFP_KERNEL);
There's lots of random spaces in this code so far and some further down too.
- struct skl_dfw_module *dfw_config = (struct skl_dfw_module *) tplg_w->priv.data;
- skl_tplg_dump_widget_info(bus->dev, dfw_config, tplg_w);
Consider implementing debugfs for this...
+/* This will be read from topology manifest, currently defined here */ +#define SKL_MAX_MCPS 30000000 +#define SKL_FW_MAX_MEM 1000000
Oh dear, this sounds like we need another ABI update to add these manifests?
- dev_dbg(bus->dev, "In%s req firmware topology bin\n", __func__);
Those "In%s" are going to come out as the function name prefixed with In. What's that for? It's just going to make the logs harder to read as far as I can tell :(
- const struct firmware *fw;
- ret = request_firmware(&fw, "dfw_sst.bin", bus->dev);
- if (fw == NULL) {
dev_err(bus->dev, "config firmware request failed with %d\n", ret);return ret;- }
We're ignoring the return value (which is what we should be paying attention to here) and checking to see if fw is NULL but fw wasn't initialized :(
- /* Index is for each config load */
- ret = snd_soc_tplg_component_load(&platform->component, &skl_tplg_ops, fw, 0);
Which index?